From Baseload to Peak: Renewables Provide a Reliable Solution

From baseload to peak: Renewables provide a reliable solution

Unless otherwise stated, this publication and material featured herein are the property of the International Renewable Energy Agency (IRENA) and are subject to copyright by IRENA.

Material in this publication may be freely used, shared, copied, reproduced, printed and/or stored, provided that all such material is clearly attributed to IRENA and bears a notation that it is subject to copyright (© IRENA), with the year of the copyright.

Material contained in this publication attributed to third parties may be subject to third-party copyright and separate terms of use and restrictions, including restrictions in relation to any commercial use.

About IRENA The International Renewable Energy Agency (IRENA) is an intergovernmental organisation that supports countries in their transition to a sustainable energy future, and serves as the principal platform for international co-operation, a centre of excellence, and a repository of policy, technology, resource and financial knowledge on renewable energy. IRENA promotes the widespread adoption and sustainable use of all forms of renewable energy, including bioenergy, geothermal, hydropower, ocean, solar and wind energy, in the pursuit of sustainable development, energy access, energy security and low-carbon economic growth and prosperity. www.irena.org

Authors: Falko Ueckerdt (Potsdam Institute for Climate Impact Research) and Ruud Kempener (IRENA)

For further information or to provide feedback, please contact: [email protected] or [email protected].

Disclaimer

This publication and the material featured herein are provided “as is”, for informational purposes.

All reasonable precautions have been taken by IRENA to verify the reliability of the material featured in this publication. Neither IRENA nor any of its officials, agents, data or other third-party content providers or licensors provides any warranty, including as to the accuracy, completeness, or fitness for a particular purpose or use of such material, or regarding the non-infringement of third-party rights, and they accept no responsibility or liability with regard to the use of this publication and the material featured therein.

The information contained herein does not necessarily represent the views of the Members of IRENA, nor is it an endorsement of any project, product or service provider. The designations employed and the presentation of material herein do not imply the expression of any opinion on the part of IRENA concerning the legal status of any region, country, territory, city or area or of its authorities, or concerning the delimitation of frontiers or boundaries. Copyright © IRENA 2015

Unless otherwise stated, this publication and material featured herein are the property of the International Renewable Energy Agency Contents (IRENA) and are subject to copyright by IRENA.

Material in this publication may be freely used, shared, copied, reproduced, printed and/or stored, provided that all such material is List of Figures ��1 clearly attributed to IRENA and bears a notation that it is subject to copyright (© IRENA), with the year of the copyright. Summary...... 2 Material contained in this publication attributed to third parties may be subject to third-party copyright and separate terms of use and restrictions, including restrictions in relation to any commercial use. 1 BA SELOAD REFLECTS DEMAND, NOT SUPPLY ...... 3

2 toDAY’S BASELOAD PLANTS HINDER FUTURE GENERATION MIX ...... 4

About IRENA 3 VARIABILI TY CHALLENGES BASELOAD PLANT CONCEPT...... 5

The International Renewable Energy Agency (IRENA) is an intergovernmental organisation that supports 4 IN TEGRATION COSTS DEPEND ON SYSTEM CONDITION ...... 8 countries in their transition to a sustainable energy future, and serves as the principal platform for international 5 FLEXIBILI TY OPTIONS MATCH VARIABLE RENEWABLES...... 10 co-operation, a centre of excellence, and a repository of policy, technology, resource and financial knowledge on renewable energy. IRENA promotes the widespread adoption and sustainable use of all forms of renewable 6 C OSTS AND BENEFITS DETERMINE ECONOMIC VIABILITY OF RENEWABLES...... 11 energy, including bioenergy, geothermal, hydropower, ocean, solar and wind energy, in the pursuit of References...... 13 sustainable development, energy access, energy security and low-carbon economic growth and prosperity. www.irena.org

List of Figures

Figure 1: the three elements of variable electricity demand: baseload, Disclaimer intermediate load and peak load...... 3 This publication and the material featured herein are provided “as is”, for informational purposes.

All reasonable precautions have been taken by IRENA to verify the reliability of the material featured in this publication. Neither IRENA Figure 2: Load, wind and solar PV temporal variations over one year for the United States and India...... 6 nor any of its officials, agents, data or other third-party content providers or licensors provides any warranty, including as to the accuracy, completeness, or fitness for a particular purpose or use of such material, or regarding the non-infringement of third-party rights, and they Figure 3: seasonal complementarity of solar PV and wind power generation in Germany accept no responsibility or liability with regard to the use of this publication and the material featured therein. based on monthly averages from January 2012 until January 2015...... 7 The information contained herein does not necessarily represent the views of the Members of IRENA, nor is it an endorsement of any project, Figure 4: generation cost data for onshore wind power, nuclear, gas product or service provider. The designations employed and the presentation of material herein do not imply the expression of any opinion on and coal power plants (with CCS)...... 10 the part of IRENA concerning the legal status of any region, country, territory, city or area or of its authorities, or concerning the delimitation of frontiers or boundaries.

Working paper 3 Summary

An oft-heard critique of renewable power generation is that renewable options are unsuitable for baseload supply, therefore fossil power and nuclear power are needed. This critique is misleading. Baseload is a demand characteristic, not a supply technology characteristic. Nuclear or coal power plants are operated in baseload mode simply because: i) they are not technically capable of operating in a more variable mode and ii) they must rely on high utilisation to recover their high investment costs.

In the future power system, the value of baseload will decrease. With higher shares of renewable power, particularly from variable sources such as wind and solar, supply and demand will be matched in a much more concerted and flexible way. Variable renewable power generation can ideally be combined with smart-grid technologies, demand response, energy storage and more flexible generation technologies, including gas power plants and dispatchable renewable power supply options. A flexible, renewables-based power system is not only reliable, but also economically efficient.

4 From baseload to peak 1 B ASELOAD REFLECTS DEMAND, NOT SUPPLY

Baseload is a characteristic of electricity In the example in Figure 1, baseload is about half peak demand and not a necessity of the supply load capacity. This illustrates that, for a typical power side system, baseload constitutes more than half of total annual electricity demand. In addition, part of the load Electricity demand (also termed load) varies over the varies over a broad range of time (peak load and inter- course of a year (Figure 1, left, shows an annual load mediate load). For example, the highest load hours are curve for Germany). In most power systems it never only recorded over a small portion of the year. drops to zero, i.e. there is a minimum load during the year, which is often termed baseload (Figure 1, right). The time in which the minimum demand, which deter- This is true for grid-connected power systems of at least mines baseload, occurs varies by power system. In North- a medium size, for two reasons. Firstly, some processes ern latitudes without air conditioning (e.g. NW Europe, continuously consume electricity. Examples include NE USA) it is typically a summer weekend night. In more industrial processes, such as aluminium smelting, or Southern latitudes where peak demand depends on air residential applications such as refrigerators, freezers conditioning (e.g. Japan, SW USA, Middle East and North and electronics in stand-by mode. The second and more Africa) it is typically a winter night. The gap between base- important reason is a result of statistics: at any moment load and peak load capacity can vary, but in the example in time, lights are switched on, mobile phones are of Figure 1 for Germany, baseload accounts for more than charging and washing machines are running. Baseload half of total electricity demand. So the question of whether is a concept that describes a characteristic of the power renewables-based power system can meet baseload is key demand side, and not a necessity of the supply side. to the viability of a renewable power supply.

Figure 1: Electricity demand (here for Germany in 2011) varies over different time frames, from hours to seasons (left). The load duration curve (right) is derived by sorting the load curve (left) in descending order. According to their duration, different parts of the load can be distinguished: baseload, intermediate load and peak load.

Load curve for one year Load duration curve maximum load peak load

Load (GW) intermediate Sorting load minimum load

baseload

1 year 1 year Time (chronological order) Time (sorted)

Source: adapted from Ueckerdt, Brecha and Luderer (2015).

Working paper 5 2 toDAY’S BASELOAD PLANTS HINDER FUTURE GENERATION MIX

Today, baseload is often covered by “baseload The objective should be to supply all parts of load, from power plants” such as nuclear or coal power baseload to peak load, in a reliable and cost-effective way. plants. However, this could hinder the future power generation mix In fact, having constant power output is not necessarily positive, but can have negative outcomes. “Base- Load is inherently variable. Therefore, a heterogeneous load power plants” actually rely on running at almost mix of different generation technologies, bringing flexi constant output throughout the year for two reasons. bility in output and incurring different degrees of fixed Firstly, these plants tend to lack the flexibility to ramp and variable costs, is more cost-effective than a single at high rates and follow variable load. More importantly, technology. Traditionally, baseload is often covered by even if baseload power plants were operated more so-called “baseload power plants”, like nuclear or coal flexible, they would still rely upon a high utilisation rate power plants. These plants are characterised by high so that enough electricity can be sold for producers to capital costs and low variable costs and, as such, prefer recover their high specific investment costs. running at a constant output. Typically, gas-combined- cycle plants are deployed for intermediate load and gas Consequently, since future power systems incorporating turbines or oil-fired plants are used for peak load.T he a higher share of renewables will require a more flexible latter “peak load plants” have low capital costs, but high interplay of its components, it cannot be guaranteed variable costs. that any technology will run at a high utilisation rate or even provide a constant output. Thus, the role of “base- Opponents of renewable power generation argue that load power plants” is likely to decrease. In fact, a large “renewables cannot supply baseload power, therefore share of “baseload power plants” could even hamper we must rely on fossil or nuclear plants”. This argument the required transformation towards renewables and is both untrue and misleading. Providing baseload power might cause ‘lock-ins’ for power systems dominated by with a single plant should not be seen as an end in itself. reliance on conventional plants.

6 From baseload to peak 3 V ARIABILITY CHALLENGES BASELOAD PLANT CONCEPT

A transformation towards variable renewables VRE cannot cover baseload power demand at all times. requires rethinking the concept of “baseload This does not need to be a disadvantage, since covering power plants” baseload power is not an end in itself. A combination of VRE and dispatchable renewable power, or of VRE There are two distinct categories of renewable power and flexible fossil-fuelled power, can reliably meet total generators: dispatchable and variable. Dispatchable power demand (including baseload) at all times. renewable power generators control their output within a specific range, just like conventional fossil power Figure 2 shows the temporal variations (hourly and plants. These generators can provide baseload pow- weekly averages) of load, and wind and solar PV er if needed. Reservoir hydropower plants, biomass production over one year for the USA and India, indexed (including biogas) power plants, geothermal power to the annual average. Hourly load varies for the USA, as plants, and concentrated solar power (CSP) plants a whole, by 50% above and below the annual average. with thermal storage (such as in molten salt) all gener- The variation of load in India is even lower than in the ate dispatchable renewable power. Integrating these USA. In contrast, for wind and solar PV power supply, renewable sources into power systems does not pose the variations are much higher than those of load. The additional challenges. In fact, many power systems al- values range from close-to-zero values up to 3‑4 times ready achieve high electricity shares from dispatchable the average annual generation. Solar PV shows periodic renewables, particularly hydropower and geothermal diurnal and seasonal cycles, while wind power shows power (e.g. in 2013: Austria 72%; Canada 61%; Colombia seasonal cycles and varies rather erratically on diurnal 79%; Iceland 100%; New Zealand 69%; and Norway time scales. The time series are derived for a future 96%). power system under the assumption that all regions of the respective country are well-interconnected. The output of variable renewable energy (VRE) sources, Consequently, variability has already been smoothened such as wind and solar photovoltaic (PV) is much less somewhat. controllable. Power generation from VRE is growing rapidly. Worldwide, newly installed capacity from wind Experiences with wind power in Spain and the UK and solar PV reached around 54 GW and 39 GW, re- confirm this range of variability. Spain had 20 GW of spectively, in 2014. This represented about 38% of global wind capacity installed in 2010, while the production of power generation capacity additions. By the end of wind varied between 1% and 76% of installed capacity 2014, there was around 370 GW of wind and 185 GW of (Martin-Martinez, et al, 2012). The minimum and solar PV power installed, together accounting for about maximum for the UK, with 10 GW of installed capacity, 10% of global power generation capacity. But this ca- were 0.05 GW and 6 GW, respectively, so production pacity is not distributed evenly; shares are much higher was 5%-60% of installed capacity (Best, 2014). Similar in some countries than in others. effects have applied to solar PV. In Germany, peak generation in summer 2013 was around 75% of installed In 2013, Denmark, Germany and Spain had renewable PV capacity, while the lowest recorded daily peak electricity generation shares of 56%, 25% and 42%, re- generation in winter 2013 was around 6% of installed spectively, with at least half of it coming from variable PV capacity (SMA, 2014). Of course the variability of renewable sources. For Denmark and Germany, electri solar PV is different from that of wind, in the sense that city trade with neighbouring countries helps to stabilise there is no solar generation at night. the grid. These examples show that it is feasible to operate power systems with high shares of variable It should be noted that variability is highest for an indi- renewable power. vidual wind or solar plant. When different plants of the

Working paper 7 Figure 2: Load, wind and solar PV temporal variations over one year for the United States and India.

USA India

1.5 1.5

1 1 Load 0.5 0.5 e) hourly values hourly values weekly mean values weekly mean values al u 0 0

0 2000 4000 6000 8000 0 2000 4000 6000 8000 e v er 5 5 w a g 4 4 o e r

v 3 3 a 2 2 1 1

Wind p 0 0

a nnu al 0 2000 4000 6000 8000 0 2000 4000 6000 8000

V 4 4 1 P (

r 3 3 2 2

Sol a 1 1 0 0 0 2000 4000 6000 8000 0 2000 4000 6000 8000 Hours of a year Hours of a year

Source: adapted from Ueckerdt et al., 2014.

same type are combined across a country or continent, ing (pumped) hydropower, biomass power generation variability decreases. Variability also decreases when dif- or energy storage can ensure power generation under ferent types of VRE such as solar and wind are combined. such conditions.

A number of countries have already shown that a mix Furthermore, the examples of Denmark, Germany and of renewables can reduce variability. Hydro and wind Spain show that up to 20-30% VRE in total annual complement each other in Brazil: in the rainy season electricity supply poses no major challenge and can be hydropower plants produce at their maximum, while in accommodated easily in power systems that are well the dry season wind generation is at its peak (IRENA interconnected with neighbouring countries. Higher and GWEC, 2012). Similarly, solar PV and wind power VRE shares pose challenges and require a rethinking generation have shown to complement each other in of power system operation and planning. With the Germany due to opposite seasonal variations. While moderate average VRE shares we are seeing already, there is a higher solar intensity and more sunny hours instantaneous penetration levels can become very high in summer, more wind is blowing during winter. Adding in some hours of the year, and VRE supply can sometimes the monthly generation averages of wind and solar even exceed electricity demand. Hence, the permanent power results in less seasonal variation than for solar PV minimum load that is covered by dispatchable power or wind alone (Figure 3). However, within each month plants is reduced and even vanishes at a certain level of the variability can still be high. For example, under the VRE deployment. In future power systems with higher conditions of a high-pressure system in Northwestern shares of VRE, distinguishing between baseload and Europe, neither wind nor a solar PV power plants can other load types, and attributing power generating operate at night. But a broader technology mix contain- technologies accordingly, is less meaningful.

8 From baseload to peak Figure 3: Seasonal complementarity of solar PV and wind power generation in Germany based on monthly averages from January 2012 until January 2015.

Average monthly power from wind and solar PV in Germany

14000

12000

10000

8000

W M 6000

4000

2000

Wind power Solar PV power

Source: Wind Journal, based on data from German transmission system operators (Amprion, Tennet TSO, TransnetBW, 50Hertz), ‘http://www. windjournal.de/erneuerbare-energie/entwicklung_windenergie_einspeisung’, (accessed May 2015).

Working paper 9 4 I NTEGRATION COSTS DEPEND ON SYSTEM CONDITION

The costs for integrating variable renewables Electricity storage can reduce profile costs, but it depend on the condition of the entire power is a relatively expensive option, starting at around system EUR 5 per kWh for pumped hydro and reaching more than EUR 150 per kWh for battery storage. There is a broad consensus that VRE creates no insur- Reducing peak demand instances through demand mountable technical barriers. However, VRE inflicts so- side management can lower this cost component called integration costs at the system level (International (IRENA, 2013). Panel on Climate Change (IPCC), 2011; Holttinen, et al., 2011; International Energy Agency (IEA), 2014; Hirth, 2. Uncertainty is caused by deviations between Ueckerdt and Edenhofer, 2015). forecasted VRE generation and actual production, which need to be balanced at short notice Integration costs are not specific to VRE. In principle, (from seconds to hours). Improved forecasting every generation technology imposes additional costs techniques have decreased associated costs, yet on the power system. However, variable renewables unpredictability remains. The impact of uncertainty have three characteristics that may require specific receives much attention in literature and public measures and additional costs to integrate these debate, yet the required flexibility is technically technologies into current power systems: temporal feasible at less than EUR 6 per MWh of VRE, less variability in VRE resource availability, uncertainty than 10% of VRE generation costs, even at high because resource availability is less predictable, and wind shares (Holttinen, et al., 2011; IEA, 2014; Hirth, the location-specific nature of resources due to their Ueckerdt and Edenhofer, 2015). geographical availability. 3. The location-specific nature of resources has an At low VRE shares, integration costs are low or even impact on investment needs in transmission and negative, which means that VRE deployment can save distribution lines. In transmission networks of costs on a system level. High VRE shares do not pose developed power systems, the resulting grid costs insurmountable technical challenges, but integration tend to be less than EUR 10 per MWh of VRE at costs will increase. wind shares of about 30%-40% (Holttinen, et al., 2011; NREL, 2012; Hirth, Ueckerdt and Edenhofer, 1. VRE provides electricity, but due to temporal 2015) – small compared to VRE generation costs. In variability VRE sources cannot be relied upon weak grids, these costs can become significant. In during peak demand times (VRE have a low so- distribution networks, distributed VRE generation called capacity credit). Thus, VRE requires some from sources such as solar PV can actually decrease dispatchable ‘back-up’ capacity in case solar and grid enhancement costs, as VRE generators, such as wind resources are unavailable. Furthermore, the rooftop solar PV stations, can be installed closer to utilisation of dispatchable power capacity is reduced, load (estimates in Europe range from EUR 2.5-5 per which increases the specific costs in the non-VRE MWh of savings). However, with higher VRE shares part of the system. These so-called profile costs, in the distribution network (typically around 10%) which comprise all costs of variability including grid enhancement costs in the distribution network back-up costs, can reach EUR 15-25 per MWh at increase. wind and solar power shares of 30%-40% (IEA, 2014; Hirth, Ueckerdt and Edenhofer, 2015). A mix of wind Note that integration costs should not be entirely attri and solar PV significantly decreases these costs. buted to VRE, because the level of grid integration costs

10 From baseload to peak necessary is highly dependent on the characteristics of Integration costs are reduced if the power system the existing power system. Integration costs will be low adapts in response to increasing VRE shares. in a VRE-friendly power system consisting of flexible Importantly, baseload power plants are not a suitable generation plants, flexible demand (including demand complement to high VRE shares. A shift from capital- side management), and strong grids. In addition, inno- intensive baseload plants to peak and intermediate vative grid operations and regulatory frameworks can load plants, which are less capital-intensive and more significantly reduce grid integration costs by harnessing flexible, can significantly reduce total costs in a system the potential for technical flexibility. with high VRE shares.

Working paper 11 5 FL EXIBILITY OPTIONS MATCH VARIABLE RENEWABLES

Variable renewables can be efficiently variations of renewable supply. In addition, pooling the combined with a flexible generation mix, supply from renewable sources distributed over large enhanced grid infrastructure, demand-side distances can significantly smoothen variability and options and energy storage decrease the need for backup capacity. Thus, further interconnecting national and regional power systems Since covering baseload power is not an end in itself, it into continental power systems is likely to decrease is not necessary to adjust and supplement VRE to cover overall energy system costs. By contrast, for island baseload power, i.e. operate at constant output. For systems costs of accommodating VRE generators tend example, storage technologies and gas power plants to be higher. A balanced mix of variable wind and solar should not be seen as an add-on to wind and solar PV PV power will further decrease costs and should be plants to provide constant generation. Instead, smart complemented with flexible generation from reservoir grid technologies (IRENA 2013), demand response, hydropower, geothermal, CSP, biomass or natural gas energy storage (IRENA, 2015a) and more flexible gen- power plants. eration technologies will be able to match supply and demand in a more concerted and flexible way (IRENA All components of a power system with high VRE and IEA-ETSAP, 2015) while “baseload power plants” shares need to complement one another. The cost will become less and less relevant for future power of a mismatch of components can be very high. systems with high VRE shares. A balanced mix of all Rapidly introducing VRE into a system that does not renewable sources is likely to help build a major pillar complement VRE well (e.g. with a large share of inflexible of future power systems, and already does so in some assets such as baseload plants or underdeveloped grid parts of the world today. infrastructure) leads to fairly high total system costs. Since power plants and transmission infrastructure take An efficient system integration of VRE requires a years to be built and last for up to half a century, transformation of the design and operation of power introducing VRE requires concerted energy planning systems. System technologies, such as enhanced today. An inadequate investment decision made today grid infrastructure, smart grid technologies, energy can hamper the transition towards renewable power storage and demand-side options, play an important generation and might even create a lock-in within a role (IRENA, 2013). Electricity demand will go from conventional-based power system. Proactive energy being variable and requiring flexibility to a source of planning, with a long-term planning that accounts flexibility in the future. Consequently, demand and for short-term variability of VRE, enables a smooth supply will become more integrated, i.e. demand-side transition towards power systems with high shares of options will be able to shift demand in response to renewable energy.

12 From baseload to peak 6 C OSTS AND BENEFITS DETERMINE ECONOMIC VIABILITY OF RENEWABLES

High shares of renewables are economically nificantly enhances the competitiveness of renewable efficient in many power systems when both energy. Many climate change mitigation studies, which costs and benefits of all power sources are consider carbon costs, show that renewables are a cru- considered cial mitigation option (IPCC, 2011; GEA, 2012). The IPCC (2011) has shown in a comprehensive review that, in the While the generation costs of renewable power projects majority of forecast scenarios, renewables become the can be in the cost range of conventional generators dominant supply option by 2050. (IRENA, 2015b), the impacts of variability might dis- courage renewable energy expansion. However, the IRENA’s REmap analysis (IRENA, 2014a) confirms that economic viability and competitiveness of VRE increase high shares of power generation from renewables ac- if the full costs and benefits of all technologies are actually reduce total power generation costs if climate counted for. change and the health impacts of conventional plants are considered. Increasing the electricity share of re- Most importantly, factoring in the appropriate costs newables in the 26 REmap countries from 18% in 2010 of climate change caused by burning fossil fuels sig- to 44% in 2030 will lead to cost savings in the range of

Figure 4: Generation cost data for onshore wind power, nuclear, gas and coal power plants (with CCS).

Comparison of levelised costs of electricity (LCOE) of wind power and other low carbon generation technologies in the UK 200

Wind power costs (including 150 integration costs at 20% wind generation share) are lower than tho se

of other low-carbon options in the UK

100 PB/MWh G

Integration costs

0 Onshore wind Proﬁle costs Increased Grid costs Onshore wind Nuclear Gas power Coal power power (including ﬂexibility (transmission power (Hinkley plant plant back-up needs and (integrated) point) (with CCS) (with CCS) costs) (balancing distribution) costs)

Source: Data for onshore wind power, and gas and coal power plants with carbon capture and storage (CCS) are based on the UK Depart- ment of Energy and Climate (DECC) calculator for low carbon scenarios (assumed discount rate of 10%) and IRENA (2015c). For nuclear power, costs are estimated from the strike price guaranteed to the operators of the planned reactor Hinkley Point C (92.5 GBP per MWh for 35 years, fully indexed to inflation). Wind integration costs are estimated conservatively according to cost values in SKM (2008), Strbac, et al. (2007), Gross, et al. (2006) and Hirth, Ueckerdt and Edenhofer, (2015). We assume wind shares of 20%. For lower shares integration costs would be reduced. Also, additional measures such as smart grid technologies, demand response, energy storage and more flexible generation technologies would reduce integration costs.

Working paper 13 EUR 5‑60 per MWh (compared to a business-as-usual sustainability impacts than renewables plants. This scenario of 26% renewable electricity in 2030). Hence, reduces the social acceptance of nuclear and CCS energy policy should account for all costs including en- plants. If a society considers sustainability concerns vironmental and health impacts when evaluating power of paramount importance, renewable energy supply options. technologies are clearly the most important low- carbon technologies. Among low-carbon technologies, renewable energy technologies take prominence for both economic and In the past, the main argument for ambitious renewa- sustainability reasons. Comparing the levelised costs of bles targets and policy support schemes was miti- electricity for onshore wind power in the UK with those gating greenhouse-gas emissions. This argument has of fossil plants combined with carbon capture and stor- broadened in recent years. Other social objectives have age (CCS) and nuclear plants implies that renewables gained importance, such as energy security, job crea- are economically favourable, even considering the cost tion, reducing local environmental damage, poverty impacts of variability (Figure 4). reduction and energy access (IPCC, 2011; IRENA, 2014b; IRENA, 2014c). There is a broad consensus that reducing In addition to the cost advantages, there are broader local environmental impact and greenhouse-gas emis- sustainability advantages of renewable energy sources sions are convincing economic arguments for a positive compared with other low-carbon technologies. Fossil cost-benefit balance of policies aimed at accelerating plants combined with carbon capture and storage the deployment of renewable energy, including ac- (CCS) and nuclear plants face much more severe counting for baseload provision.

14 From baseload to peak References

Best, C. (2014), “Wind power results 2013/14”, http:// IRENA (2014c), “Rethinking Energy”, IRENA, Abu Dhabi, clivebest.com/blog/?p=6259, accessed May 2015. September 2014, www.irena.org/rethinking. Gross, R., et al. (2006), “The costs and impacts of IRENA (2015a), “Battery storage for renewables; market intermittency: an assessment of the evidence on the status and technology outlook”, IRENA, Abu Dhabi, costs and impacts of intermittent generation on the January 2015, www.irena.org/publications. British electricity network”, Energy Research Centre, IRENA (2015b), “Renewable power generation costs”, London, http://www.uwig.org/0604_Intermittency_ IRENA, Abu Dhabi, January 2015, www.irena.org/costing. report_final.pdf. Gunnar, L., et al. (2013), “Economic Mitigation IRENA – IEA-ETSAP (Energy Technology Systems Challenges: How Further Delay Closes the Door for Analysis Program) (2015) “Renewable energy integration Achieving Climate Targets”, Environmental Research technology brief”, IRENA, Abu Dhabi, January 2015, Letters, Vol. 8, No. 3, pp. 034033, doi:10.1088/1748- www.irena.org/publications. 9326/8/3/034033. Martin-Martinez, S., et al. (2012), “Participation of wind Hirth L, F. Ueckerdt, and O. Edenhofer (2015), power plants in the Spanish power system during “Integration costs revisited – An economic framework events”, Power and Energy Society General Meeting, for wind and solar variability”, Renewable Energy, 2012 IEEE, 22-26 July 2012. Vol. 74, pp. 925-939. NREL (National Renewable Energy Laboratory) (2012), Holttinen, H., et al. (2011), “Impacts of Large Amounts “Renewable Electricity Futures Study”, NREL/TP- of Wind Power on Design and Operation of Power 6A20-52409, Golden CO, www.nrel.gov/analysis/re_ Systems, Results of IEA Collaboration”, Wind Energy, futures. Vol. 14, No. 2, pp. 179–92. SKM (Sinclair Knight Merz) (2008), “Growth Scenarios IEA (International Energy Agency) (2014), “The Power for UK Renewables Generation and Implications for of Transformation – Wind, Sun and the Economics of Future Developments and Operation of Electricity Flexible Power Systems”, IEA/OECD (Organisation for Networks”, BERR Publication, URN 08/1021, http:// Economic Co-operation and Development), Paris. webarchive.nationalarchives.gov.uk/+/http:/www.berr. IPCC (Intergovernmental Panel on Climate Change) gov.uk/files/file46772.pdf. (2011), “Special Report on Renewable Energy Sources SMA (2014), “Performance of photovoltaics (PV) in and Climate Change Mitigation”, O. Edenhofer, et al., Germany”, www.sma.de/en/company/pv-electricity- (eds.), Cambridge University Press, Cambridge and produced-in-germany.html. New York. IRENA (International Renewable Energy Agency) and Strbac G., et al. (2007), “Impact of wind generation on GWEC (Global Wind Energy Council) (2012), “IRENA- the operation and development of the UK electricity GWEC: 30 Years of Policies for Wind Energy”, IRENA, systems”, Electric Power System Research, Vol. 77 pp. Abu Dhabi, www.irena.org/publications. 1214-27. IRENA (International Renewable Energy Agency) (2013), Ueckerdt, F., R. Brecha and G. Luderer (2015), “Analyzing “Smart grids and renewables: a guide for effective major challenges of wind and solar variability in power deployment”, IRENA, Abu Dhabi, November 2013, www. systems”, Renewable Energy, vol. 81, pp. 1-10, September irena.org/publications. 2015. IRENA (2014a), “REmap 2030: A Renewable Energy Ueckerdt, F., et al. (2014), “Representing variability of Roadmap”, IRENA, Abu Dhabi, June 2014, www.irena. wind and solar in integrated assessment models by org/remap. using residual load duration curves, Poster at Seventh IRENA (2014b), “Renewable Energy and Jobs – Annual Annual Meeting of the Integrated Assessment Modeling Review”, IRENA, Abu Dhabi, January 2014, www.irena. Consortium (IAMC) 2014, Washington D.C., 17-19 org/jobs. November 2014.

Working paper 15 www.irena.org